Lightnews — Scholar-powered news

Christoph Riedl @criedl.bsky.social · 14h

Emergent Coordination in Multi-Agent Language Models

When are multi-agent LLM systems merely a collection of individual agents versus an integrated collective with higher-order structure? We introduce an information-theoretic framework to test -- in a p...

arxiv.org

Christoph Riedl @criedl.bsky.social · 14h

Emergent properties have functional benefits and realign internal group structure. Results mirror collective intelligence principles: effective performance requires both alignment on shared objectives and complementary contributions across members.

1 2

Christoph Riedl @criedl.bsky.social · 14h

We find multi-agent systems have capacity for emergence - they are real "teams" that are more than the sum of their parts. And we can steer them with clever prompts. The ToM condition in particular leads to stable specialization and goal-directed complementarity across agents

1 1

Christoph Riedl @criedl.bsky.social · 14h

Experiments use a simple guessing game without direct agent communication and only minimal group-level feedback with three randomized interventions: plain, agents with personas, and personas with an instruction to "think about what other agents might do" (a ToM prompt)

1

Christoph Riedl @criedl.bsky.social · 14h

When are multi-agent LLM systems merely a collection of individual agents versus an integrated collective with higher-order structure? New paper shows multi-agent LLM systems have capacity for emergent coordination and how to steer them ...

1

Reposted by Christoph Riedl

NetSci 2026 @netsciconf.bsky.social · 8d

📌 Save the Date!

The flagship conference of the Network Science Society - 𝗡𝗲𝘁𝗦𝗰𝗶 𝟮𝟬𝟮𝟲 - is coming to Northeastern University’s Network Science Institute, 𝗝𝘂𝗻𝗲 𝟭-𝟱, 𝟮𝟬𝟮𝟲.
Prepare to share ideas, discoveries & challenges in network science.

Registration opens soon! 🔗 www.netsci2026.com

34 40

Reposted by Christoph Riedl

Ethan Mollick @emollick.bsky.social · 15d

Some important findings in this paper:
1) Working with AI boosts the performance of people solving math, science & ethics questions
2) The biggest boost is for the hardest problems
3) High performers remain highest performing, but low performers gain more
4) People who are good with AI gain most

2 24 97

Christoph Riedl @criedl.bsky.social · 15d

osf.io/preprints/ps...

OSF

osf.io

Christoph Riedl @criedl.bsky.social · 15d

New paper on quantifying human-AI synergy in a large dataset with AI-alone, human-alone, and human-AI together.

1

Christoph Riedl @criedl.bsky.social · 15d

aka people with higher theory of mind write better prompts and get better answers from AI. AI on the other hand is quite bad responding to moment-to-moment fluctuations in ToM

1

Christoph Riedl @criedl.bsky.social · 15d

Crucially we find that solo-ability and collaboration-ability are distinct traits. And it turns out social skills (theory of mind) predict AI collaboration ability AND AI response quality.

1

Christoph Riedl @criedl.bsky.social · 15d

Who benefits from AI and on which task? And how can we measure human-AI synergy?

1 1

Christoph Riedl @criedl.bsky.social · Jul 30

Rivalry can backfire! We study the role of peer effects on top of known economic effects like competition pubsonline.informs.org/doi/abs/10.1...

Reposted by Christoph Riedl

David Lazer @davidlazer.bsky.social · Jul 10

Very pleased our paper looking at the diffusion of complex contagions is out in @pnas.org www.pnas.org/doi/10.1073/...

Diffusion of complex contagions is shaped by a trade-off between reach and reinforcement | PNAS

How does social network structure amplify or stifle behavior diffusion? Existing theory suggests that when social reinforcement makes the adoption ...

www.pnas.org

3 14 34